Repository navigation
Wire batch-norm fusion into CpuBackend.preprocess (#24193) - #24193
Merged
Merged
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/24193
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New Failures, 50 PendingAs of commit 63d563f with merge base 5723674 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
Contributor
|
@JakeStevens has exported this pull request. If you are a Meta employee, you can view the originating Diff in D126784606. |
digantdesai
approved these changes
Oct 27, 2026
digantdesai
left a comment
digantdesai
left a comment
Contributor
There was a problem hiding this comment.
Review automatically exported from Phabricator review in Meta.
This PR needs a
|
meta-codesync
Bot
changed the title
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
force-pushed
the
export-D126784606
branch
from
October 27, 2026 15:31
3ea6045 to
0f5523a
Compare
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
Summary: CPU partitioning and delegate I/O required every tensor to be FP32, leaving indices, masks and integer outputs outside the delegate. Accept static uint8, int8, int16, int32, int64 and bool tensors, and validate runtime I/O against the serialized dtype. Existing allocation and ET kernel routes preserve the values without conversion through FP32. Integer delegation also exposes `index.Tensor`/`index_put`, whose indices are `Tensor?[]`. Native serialization emits a plain tensor list when every index is present, but ET's kernels unbox a list of optional tensors. Mark those Reference routes with `optional_tensor_lists` so the CPU runtime binds optional lists for them; the shared serializer and the other Native engines are unchanged. Add exact runtime regressions for integer precision, indices, mixed outputs, constants and optional index lists. Dynamic shapes/scalars and quantized execution remain separate. The Native schema and storage contract do not change. Authored with AI assistance using Codex and Claude Code. Reviewed By: digantdesai Differential Revision: D123332329
Summary: Adds `FuseBatchNormWithLinearPass` to `backends/transforms`, the linear counterpart of `FuseBatchNormWithConvPass`. It folds an inference batch norm into the preceding `aten.linear` weight and bias using `torch.nn.utils.fusion.fuse_linear_bn_weights`, so backends can drop the BN without serializing extra weights. Fusion only happens when it is numerically equivalent: the linear output must be rank 2 (the only case where BN's channel dim is linear's output dim), every weight/bias/running-stat input must be a constant parameter, BN must only be consumed through `getitem(0)`, and training-mode `native_batch_norm` is skipped. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D125532471
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
force-pushed
the
export-D126784606
branch
2 times, most recently
from
October 27, 2026 16:08
77f8a3b to
42720b9
Compare
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
added a commit
to JakeStevens/executorch
that referenced
this pull request
Oct 27, 2026
Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
JakeStevens
added a commit
that referenced
this pull request
Oct 27, 2026
`backends/transforms/targets.bzl` defines the `fuse_batch_norm_with_linear` `python_library` twice (lines 138 and 152). The first copy comes from #24192. The second came in with the squash of #24193, whose branch picked up the block again when it was merged with `main`. Buck refuses to register the same target twice, so any command that loads `//backends/transforms/...` fails: ``` error: Attempted to register target root//backends/transforms:fuse_batch_norm_with_linear twice ``` This breaks `unittest-buck` on `main` and on open PRs. This change deletes the second, identical definition. No other targets in the file are duplicated. ### Test plan - Checked that `targets.bzl` now defines each target name once. - CI: `unittest-buck` should get past the `buck2 query` step.
Erik-Lundell
pushed a commit
to Erik-Lundell/executorch
that referenced
this pull request
Oct 28, 2026
…torch#24193) Summary: Run the shared `FuseBatchNormWithConvPass` and `FuseBatchNormWithLinearPass` in `CpuBackend.preprocess`. Unfused `_native_batch_norm_legit_no_training` already executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed in `CPU_PREPROCESS_PASSES` (mirroring `XNNPACKPassManager.passes`) so every lowering flow gets them from one extension point. Fusion introduces `get_attr` constants, so preprocess follows with `lift_constant_tensor_pass` (the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones. Authored with Claude Code. Reviewed By: digantdesai Differential Revision: D126784606
Erik-Lundell
pushed a commit
to Erik-Lundell/executorch
that referenced
this pull request
Oct 28, 2026
`backends/transforms/targets.bzl` defines the `fuse_batch_norm_with_linear` `python_library` twice (lines 138 and 152). The first copy comes from pytorch#24192. The second came in with the squash of pytorch#24193, whose branch picked up the block again when it was merged with `main`. Buck refuses to register the same target twice, so any command that loads `//backends/transforms/...` fails: ``` error: Attempted to register target root//backends/transforms:fuse_batch_norm_with_linear twice ``` This breaks `unittest-buck` on `main` and on open PRs. This change deletes the second, identical definition. No other targets in the file are duplicated. ### Test plan - Checked that `targets.bzl` now defines each target name once. - CI: `unittest-buck` should get past the `buck2 query` step.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary:
Run the shared
FuseBatchNormWithConvPassandFuseBatchNormWithLinearPassinCpuBackend.preprocess. Unfused_native_batch_norm_legit_no_trainingalready executes through the reference provider, but as a separate kernel with its own serialized weight and stats; folding it into the preceding conv/linear removes that kernel from the delegate. The passes are listed inCPU_PREPROCESS_PASSES(mirroringXNNPACKPassManager.passes) so every lowering flow gets them from one extension point.Fusion introduces
get_attrconstants, so preprocess follows withlift_constant_tensor_pass(the native serializer rejects unlifted constants), then drops parameter/buffer/constant inputs left without users. Without that step the delegate serialized the original weights and batch norm stats next to the fused ones.Authored with Claude Code.
Reviewed By: digantdesai
Differential Revision: D126784606